* StreamInfo describe stream. * - sampleRate is the sampleRate of original signal * - sampleStep is the gap between consecutive frames in number

A DataFlow object hold a directed acyclic graph of computational steps describing how to compute some audio features. DataFl

A AudioFileProcessor object allow to extract features from audio files, and possibly write output features into files. It mu

A Engine object is in charge of processing computations defined in a :py:class:`DataFlow` object on given inputs. .. doctest

FeaturePlan is a collection of features to extract, configured for a specific sample rate. :param sample_rate: analysis samp

Tremelo and Grain description, according to [SE2005]_ and [AE2001]_. AmplitudeModulation uses :class:`Envelope` to describe tremolo and grai

classdocs

Compute autocorrelation coefficients *ac* on each frames. .. math:: ac(k) = \sum_{i=0}^{N-k-1}x(i)x(i+k)

Feature transform that compute peaks of the autocorrelation function, outputs peaks and amplitude.

Compute the beat histogram according to [GT2002]_, but using :class:`OnsetDetectionFunction` as onset detection function. .. [GT2002] Ge

Compute the Constant-Q transform according to [CS2010]_ with improvements from [JPCQT]_. .. [CS2010] C.Schörkhuber and A.Klapuri, *CONST

Compute the Constant-Q transform according to Blankertz's implementation [BB]_, with improvments from [JP2010]_. .. [BB] B.Blankertz

Feature transform that compute cepstrum coefficients of input feature frames. (use DCT II) .. math:: cep = dct(log(x))

Chords recognize chords from chromagrams, according to L.Oudre's algorithm [LO2011]_. .. [LO2011] Oudre, L. and Grenier, Y. and Fevotte, C.,

Chroma compute short-term chromagram according to [BP2005]_. .. [BP2005] Bello, J.P. and Pickens, J. *A Robust Mid-level Representation for

Chroma2 compute short-term pitch profile according to [ZK2006]_. .. [ZK2006] Y. Zhu and M.S. Kankanhalli. *Precise pitch profile feature ext

Compute onset detection using a complex domain spectral flux method [CD2003]_. .. [CD2003] C.Duxbury et al., *Complex domain onset detec

* The Component class represent a computation step. The 4 main methods are * init, reset, process, flush. Theses methods are called according to th

* DataBlock manage a memory block used to store numerical data streams. * It's usually created by an OutputBuffer and filled with data. Then can

Compute temporal derivative of input feature. The derivative is approximated by an orthogonal polynomial fit over a finite length window.

Compute energy as root mean square of an audio Frame. .. math:: en = \sqrt\\frac{\sum_{i=0}^{N-1}x(i)^2}{N}

Extract amplitude envelope using hilbert transform, low-pass filtering and decimation.

Centroid, spread, skewness and kurtosis of each frame's amplitude envelope. For more details about moments, see :ref:`Shape Statistics <s

Segment input signal into frames. First frame has zeros on left half so that it is centered on time 0s, then consecutive frames are equa

Feature transform that compute histogram of input values

* InputBuffer represents a input data stream where a component * can read data. * * Here's some classical way to deal with InputBuffer *

Compute the Linear Predictor Coefficients (LPC) of a signal frame. It uses autocorrelation and Levinson-Durbin algorithm. see [JM1975]_.